Document 3exk3b5J90KBQkKk2ejXVLknn

Scand. j. work environ. & health 4 (1978): suppL 2, 229--284 Mutagenesis, teratogenesis and carcinogenesis in man FRENTZEL-BEYME, R., THIESS, A. M. and WIELAND, R. Survey of mortality among employees engaged in the manufacture of styrene and polystyrene at the BASF Ludwigshafen works MEINHARDT, T. J., YOUNG. R. J. and HARTLE, R. W. Epide miologic investigations of styrene-butadiene rubber produc tion and reinforced plastics production ................................ NICHOLSON, W. J., SELIKOFF, I. J. and SEiDMAN, H. Mortality experience of styrene-polystyrene polymerization workers: Initial findings ........................................................................ HOLMBERG, P. C. Two children with central nervous defects born to mothers exposed to styrene at work........................ FLEIG, I. and THIESS, A. M. Mutagenicity study of workers employed in the styrene and polystyrene processing and manufacturing industry ......................................................... MERETOJA, T. (t). JARVENTAUS, H., SORSA, M. and VAINIO, H. Chromosome aberrations in lymphocytes of workers expo sed to styrene ........................................................................ 231 240 247 253 254 259 . -ii j -- i*? i~^'~ ' j^~H~ ~ ~ ~ .. ` ir' zf* sssxRii - ' IrEfi' utiridum ^ --/ 229 ii - UCC 057499 Scand. }, work environ, & health 4 (1978): suppl. 2, 231--239 Survey of mortality among employees engaged in the manufacture of styrene and polystyrene at the BASF Ludwigshafen works by R. FRENTZEL-BEYME,1 A. M. THIESS * and R. WIELAND * FRENTZEL-BEYME, R., THIESS, A. M. and WIELAND, R. Survey of r orfaUty among employees engaged in the manufacture of styrene and polystyren- at the BASF Ludwigshafen works. Scand. j. work environ, dr health 4 (1978): .uppL 2, 231--239. A mortality survey of 1,960 employees currently or previously e..posed to styrene at work has shown that the observed number of deaths from cancer or other causes is no higher than the number expected in the German population or in a group of BASF employees. Analyses of deaths occurring after a minimum period of five years in groups exposed for different durations (5 years, 10 years, 13 years, more than 15 years) indicated no increase in mortality with exposure time. This latter observation also applied to malignant tumors; therefore it may be inferred that prolonged exposure to styrene is not a cancer hazard. Key words; cancer, chemical industrial workers, mortality, prospective study, styrene. A survey of mortality among persons that .German employees. The follow-up was have been exposed to styrene during their more difficult in the case of former non- employment in the BASF Aktiengesell- German employees, many of whom had schaft has been carried out in order to returned home to countries without discover whether there was an excess registration systems comparable to those incidence of malignant tumors. The cohort included ail past and present employees engaged in the manufacture of styrene or polystyrene for more than one month since production started in 1931. The date when exposure started, the date of birth, the date of leaving the plant, Table 2. Numbers of person-years of observa tion for the styrene cohort (1,960 persons) and the vinyl chloride cohort (1,681 persons). and the reason for leaving the plant were extracted from the plant records of each Group Total period of observation for group employee or exemployee and recorded. When a person was no longer engaged Styrene Vinyl chloride in styrene or polystyrene production, a 15-- 24 follow-up procedure was carried out; this 25-- 34 was successful for 93 /# of all formeri * 335-- 44 45-- 54 55-- 64 i Deutsches Krebaforschungsxentrum, Heidel 65-- 74 berg, Federal Republic of Germany. 75-- 84 3 Occupational Medicine and Health Pro 85--100 tection, BASF Aktiengesellschaft, Ludwigshafen/Rhein, Federal Republic of Germany. Total 2,732.9 8,819.4 5.532.4 3,407.7 1,379.5 258.4 7.8 0.0 20,138.1 3,156.08 6.718.35 5,809.79 2^92.14 1,175.00 194.77 20.94 0.00 19.767.07 231 In Germany, and it was successfully carried out for only 29 */ of the personnel, who however had been exposed in almost the half of all cases for less than six months. The technical data and the respective recordings of workplace concentrations have been reported in another presen tation in these proceedings (4). It was not possible to ascertain how long an individual employee had been exposed to lower or higher levels of styrene, be cause most employees had carried out various tasks within the plants. However, since both the equipment and safety pre cautions had been considerably improved over the years (for instance, now there are only closed systems), the cohort was Tabic 2. Causes of death in the styrene cohort (1,960 persons) a and comparisons of tno4dt""n with expectancies. Serial mim- ber ICO ber 0b_ strata Expected Rhlnehemia-PaJatlnst* 1*70--1975 Ludwlgihafen 1*70--1175 P A|# group (O/S) P P A* poup (O/l) P 1 on Tuberculofls i 1.048 1.180 2 0--134 Infection* and paraatte* 2 1-538 0.455 1.840 0.490 3 151 Stomach cancer 2 2-58* 2272 4 i s 153 Colon cancer 1 154 Rectal cancer 1 157 Pancreatic cancer 2 1201 30--34 (1/0.42) 1.010 0.721 0.163 0541 1.1*0 30--34 (1/0.0) 0274 0213 0.700 0.157 02 7 * 15t sPdeiirMitoianeal ear- 1SS IS* MkL^aantneo- 1 0.141 0.132 40--14 (1/0512) 0.013 0220 0202 40--44 (1/0.0) 3 5.44S 5.720 02 plaam of the urinary system 10 202 Splenic tumor 11 250 Diabetes 12 252--279 Other metabolic dlieaeec 13 400 Malitnant hyper- tension 14 410 Acute rayocar- 1 1 1 1 1 053* 0.0S7 1.014 0207 0.171 0.416 0.063 0.1ST 0.151 50--54 (1/0.014) 55--50 (1/0.010) 30--34 (1/0.0) 0.014 0520 05 0.415 0.058 0.954 0223 0.130 0240 0.057 0210 0.190 0.127 50-34 (1/02) 55--59 (1/0.084) 50-34 (1 0.0) 92 0.014 00 elal Infarction 13 13.71* 15 411 Acute lachemic 13.712 heart rtlaeeie 1 0277 0214 50--54 (1/0.044) 0.043 0.492 0210 1* 413 Chronic lachemic heart dlieiae 1 1218 2.0 17 420--427 Other heart difeaaee 2 1530 0.452 1.442 0.423 50-54 (1 0.023) 003; It 42S Myocardial dlaeaaee 2 1.930 0575 1.174 0223 19 432--435 Cerebrovascular diaeaaea 1 20 436 Apoplexy 4 21 440 Arterioacleroilt i . 21 490--491 Bronchitli 1 23 493 Asthma 1 24 317--519 Pneumonopathy i 25 571 Hepatic cirrhotlt 5 26 572--373 Other liver diaeaaea 1 27 377 Pancreatopathy 1 2S 795--790 Uncertain or unknown 3 ! 800--807 Railroad accident* 1 30 810--819 Automobile accident* 7 31 820--827 Accidents with vehicles not in traffic 1 32 910--929 Other accident* 3 33 953 Suicide 3 34 960--989 Homleid* l 35 980--989 Woundinc, poa- aibly Intentional 1 0.275 3.420 0.263 1.049 0.671 0242 6.653 0231 0.424 0.533 0.290 9.440 0240 0.446 0.231 0 489 0215 0282 0248 0.017 0.252 50--54 (1/0,052) 55^*58 (1/0.033) 40 -44 (1/0.025) 30--34 (2 0.025) 50--54 (1/0.011) 0.051 0.033 0.025 0.0 0011 0.295 2.908 0.315 1.239 0.550 0.200 5.920 0.339 0.306 0.754 0.179 7.992 0.256 0232 0270 0.423 0.151 0281 0.264 0.041 0.164 50--54 (1 0.059) 55-59 a 02241 40-44 (1 0.036) 30--34 (1. 0.049) 30--34 (2 0.045) 50-54 (1 0.0) O.fltf 0024 0 09* 0.041 0001 00 0.302 2.199 2.165 0513 0.261 0277 0268 0.401 20--24 (1 0.020) 40--44 (2/0242) 0.163 0.14* 50-54 (1/0.000) 0.020 0.047 0.000 0.360 0229 20--24 (1 02) 2.297 0.403 2.014 0227 40--14 (2 0246) 0246 0.421 0.055 0.053 50--54 (1/0.0) 0.0 O.Kf 02 34 0--79S Natural cauaes 5# 742SS 71272 37 800--99* Unnatural cauaes 17 22.096 19.74* 3S 0--99* All cauae* 73 96.465 91.120 * Total period of observation 20.13* yean. 232 Tat div exp not pre bet An thi. pre wo ch anc foil I enc fur pic ves ye; ve va: be* 1. obi pec sta the Lu ex; are nu. the ch: sur - .swj: JS.r-brv; vi':*---- IcejcV* 057501 li -- - n eiianr - -i -- ifif*--- -m i huu long ?n^Bosed yrlro, be amed out However, afety pre improved iow there ohort was icidences with 970--1975 (O/E) P a o.o) 0.0 1 0.0) 0.0 1 0.0) 0.0 rf 0.024 0.0 0 036) 0.035 0 059) 0.037 0 034) 0.024 0 036) 0 049) 0 0451 00) 0.036 0.041 0.001 0.0 0 0) 0 248) 0.0 0.026 0,0) 0.0 Table 3. Incidence of deaths in the styrene cohort by age and year. Afe (years) Number of deaths 1955--1950 1951--1955 1955--1970 15--24 25--34 35-44 45--54 55--54 55--74 75--54 Totals N V. 110 032 1 15 035 1 2 11 004 000 3 10 27 4.1 13.5 36.5 One of the dead had been exposed for 15 d only. 1971--1976 3 4 2 8 8 7 2 34 45ja Total for ago N /. 9 5.7 9 12 9 12.2 16 21.6 22 29.7 11 14J 2 2.7 74 _ -- 100.0 divided into groups of those who had been exposed before 1960 and those who had not been exposed before 1960. This procedure gave a crude differentiation between possible levels of exposure. Another reason for dividing the cohort in this way was the comparatively high proportion of non-German employees working in the plant after 1960; this changed the composition of the cohort and, as mentioned before, reduced the follow-up rate considerably. For the analysis of the data with refer ence to period of exposure, the cohort was further subdivided into groups of em ployees exposed for periods of up to 5 years, 5 years or more but less than 10 years, 10 years or more but less than 15 years, and 15 years or more. The total number of years of obser vation for the 1,960 exposed persons has been broken down by age groups in table 1. Table 2 gives the number of deaths observed, by causes of death, and the ex pected numbers based on the 1970--1975 statistics for Rhinehessia-Palatinate and the 1970--1975 statistics for the town of Ludwigshafen. Where appropriate, the expectancies for the pertinent age group are given, too. Whenever the observed number of deaths exceeded expectancy, the probability of such a deviation by chance was determined, under the as sumption of the Poisson distribution. Another basis of comparison was pro vided by the use of an internal reference group consisting of a cohort of similar composition and size that resulted from studies on exposure to vinyl chloride. Data relating to mortality in this group are being published elsewhere. The total numbers of years of observation and the breakdown by age of this cohort are given in table 1. RESULTS General mortality Table 3 reveals the breakdown of the 74 deaths by age and year of death. Mortality was highest in the age group 55--64 years. Table 4 shows the proportional break down by causes of all deaths observed in the cohort and the corresponding distri bution for the whole population of the Federal Republic of Germany. When the two sets of relative frequencies of partic ular causes of death are compared, it is noticeable that within the cohort of those exposed to styrene the proportion of deaths due to malignant neoplasms was less than in the country as a whole, namely, 14.9 / compared with 18.5 /#. Similarly, the proportion of deaths due to cardiovascular diseases was appreciably less in the cohort than in the country as a whole. The proportion of deaths due to 233 ucc 057502 Table 4. Cause of death in the styrene cohort compared with causes of death in the Federal Republic of Germany, 1968--1972. ICD Cause of death Styrene cohort N /. 000--138 Infections & parasites (respiratory tuberculosis) 140--199 Malignant neoplasms 200--209 Neoplasms of lymphatic and hematogenic systems 240--279 Metabolic and nutritional diseases -90--498 Cardiovascular diseases 480--919 Diseases of the respiratory tract 520--577 Diseases of the digestive tract (hepatic cirrhosis) 950--959 Suicide 960--999 Other violent deaths 780--796 Unknown, unexplained causes 800--949 Accidents (traffic) (others) 280--289 290--389 1[ Other causes 580--629 J 680--738 000--999 Total 2 (2) 11 1 2 27 3 7 (3) 3 l 4 13 (9) (4) 0 74 One of the dead had been exposed for 19 d only. b PMR " proportional mortality ratio. 2.7 (2.7) 14.9 U 2.7 36.5 4.1 9.4 (6.7) 4.1 1.3 5.4 17.6 (12.2) (5.4) 0.0 100.0 Federal Republic of Germany */ PMR< 1.5 (0.9) 18.5 180 (300) 81 1.2 108 2.0 136 43.2 84 9.0 46 6.1 154 (2.7) (248) 2.2 186 0.3 434 3.8 142 6.3 279 (3.6) (339) (2.7) (260) 6.3 0 100.0 -- ****j1! diseases of the liver and other digestive The 56 deaths from natural causes organs was, however, higher in the cohort within the cohort were considerably than in the rest of the country, but this fewer than the numbers expected for should perhaps be considered in the light populations with the same age structure, of the generally high incidence of deaths namely, 71 (based on Ludwigshafen due to hepatic cirrhosis in Khinehessia- statistics) and 74 (based on statistics for Palatinate (table 2) -- where most Ger the administrative region Rhinehessia- man wine is produced. Palatinate). There was, however, good .-.'Si The fact that the proportion of acciden agreement between observed and expected tal deaths was nearly three times as large figures for specific causes of death when a-. in the cohort as in the total population the expectancies were based on the total points to the uselessness of comparisons number of years of observation for the of this sort. The age structures of the two whole cohort. Certain events were ob populations are simply too different For served that would have occurred very this reason, the expectancies for particular rarely in the general population, had this age groups were calculated (making use of been observed over the whole period of `'P- the total number of years of observation time. These causes of death were: peri m'v. for each group) and compared with the toneal carcinosis in the age group 46--44 numbers of deaths observed, as shown in years; colon-rectum cancer in the age i table 2. group 30--34 years; pancreatic tumor in a. 234 r- A'n.-. ' EeuiVMin fm 7-a-1-'*-'--..- ---- w--if|| p*i 'HryWM inwiSfcMfcysiasaKiGiisft UCC 057503 the age group 50--54 years; metabolic disease in the age group 55--59 years; malignant hypertension in the age group 30--34 yean; and disease of the liver in the age group 40--44 years. Since, however, none of these events occurred more than once in the whole cohort, they cannot be regarded as in* dicative of specific hazards involved in exposure to styrene. The causes/of three deaths could not be determined in spite of efforts made by the authorities. Mortality in relationship to period of exposure In accordance with a suggestion made by Fox (3), the cohort was modified by the elimination of all those who had been under observation for less than five years from the time of first exposure and the Table 5. Deaths in the reduced styrene cohort (members observed at least 5 years): Comparison of observed and expected deaths by duration of exposure. Exposure Exposure (years) started Natural deaths N Ne P Unnatural deaths N Ne P All deaths N NE P 0-- 4 5-- 9 10--14 >15 Total pre 1960 post 1960 pre 1960 post 1960 pre 1960 post 1960 pre 1960 post I960 pre 1960 post 1960 6 11.291 8 18.126 8 9.960 8 3.105 4 8.589 1 0.128 13 9.316 31 37.156 17 21.36 -- -- 0.014 -- 0.12 0.148 -- 1 3.342 0 5.187 1 1.657 3 0.677 3 0.877 0 0.027 1 0.716 6 6.591 3 5.891 -- -- 0.031 0.059 -- 0.48 -- 7 14.633 8 23.313 9 11.616 11 3.782 7 7.466 1 0.155 14 10.032 37 43.747 20 27.25 -- -- 0.002 0.144 0.137 -- The expected deaths are based on the reference population mortality of Rhinehessia-Palatlnate in 1972--1975. Table 6. Deaths in the reduced styrene cohort (members observed at least 5 years): Comparison of observed and expected deaths by age groups. Age (years) Exposure started _ Natural deaths N Ne P Unnatural deaths N Ne P All deaths N Ne p 15--24 25---34 35--44 45--54 55--64 65--74 75-- Total pre 1960 post 1960 pre 1960 post 1960 pre 1960 post 1960 pre 1960 post 1960 pre 1960 post 1960 pre 1960 post 1960 pre 1960 post 1960 pre 1960 post 1960 0-- -- -- 3 1.476 3 4.015 0 7 8.814 4 6.779 14 14.581 6 6.361 6 7.976 4 2.954 1 0.524 -- 31 37.156 17 21.36 -- 0.19 -- 0.342 0.408 1 0.387 1 1.565 0 3 1.902 1 1.786 1 1.494 1 0.74 1 1.066 0 0 0 0 -- 6 6.591 3 5.891 0.321 0.297 0.523 1 0.495 1 2.731 3 3.458 6 5.917 1 5.468 8 10.309 5 7.947 15 15.647 6 6.83 6 82140 4 3.053 1 0.537 -- 37 43.747 20 27.25 0.365 0.415 The expected deaths are based on the reference population mortality of Rhinehessia-Palatlnate in 1972--1975. 235 ucc 057504 Table 7. Details of fatal cases of cancer in the styrene cohort. Study num ber sit of tumor Diagnosis Date end kind of diagnosis Date of death Ago at time of (years) Exposure started Time Time ex* Qbpoead eaevad month (years) 21250 Tjl"f 21055 Lungs 2O0C3 Lungs Carcinoma: motaatadzlDg bronchogenic, dlaerata. 23 Sept 74 1* Sept 74 49 30 April 70 keratinized squamous cello; many mataereaa*. ex. hi kidneys, bonce, and lymph nodai 2 Puloionwy s oet a 29 Oet SB radiological 50 5 M. 53 3 Pleurltte cardnonatom acoortiperiylng bronchial car- 24 r*b. 73 74 1 Dae. 00 <7 4 7 15 21003 Stomach Tubular nardnoma of the stomach and hepatic drrhoati 31 May 71 7 June 71 hleiotogleil 55 5 Sept 50 S3 212S6 Stomach CUrrlnoma of the stomach, wtth metaitaaM, partly polypous. partly diffusely Infiltrating 23 April 55 9 June 65 histological 60 9 March 39 322 21244 Pancreas According to case records', uncertain primary tumor, metaitaaM In lumbar ver tebrae and pancreas 11 las 67 54 1 Oct 52 171 21157 Peritoneum, Atmold of the colon Peritoneal caretnoeii; adenocarcinoma of the tigma (colon/rectum) with multiple hepatic metaifaeei, pars-aortal lymphatic tnetaitaaei. wide-spread adhe sion of the peritoneum to the primary tumor 23 July <3 histological 3 Jan. 64 44 1 Aug. 55 32 20427 Rectum Adenocarcinoma of the rectum; denie Infiltration of the Intestinal wall with adenoid structured 24 July 59 histological biopsy 3 Aug. 69 61 13 July 63 72 209*7 Probably kidney Site of primary tumor not known: lilac crest, appears as metastasis of a hypernephroid carcinoma histological 2* June 65 53 20 Oct 49 183 21370 Sifma Infiltrating adenocarcino ma of the sigma (coloni; ileus accompanying perito neal carcinosis: recurrent tumor penetrat ing into the bladder 20 Jan. 72 histological 24 Oct 72 laparotomy 24 Oct 73 biopsy 25 Oct 72 34 * July 65 53 21292 Spleen Splenic tumor: no histolog ical examination! Uncertain dignity 5 Nov. 69 54 IT Aug. 59 59 21689 Head of panerta* No histological diagnosis: death certificate 2 March 70 66 28 Aug. 64 47 U 27 IS * 5 13 7 10 6 # subdivision of the remainder into groups Table 6 shows the observed and expect --iv, of people that had been exposed for less ed numbers of deaths for the cohorts with than 5 years, 5--9 years, 10--14 years, and periods of employment before and after 15 years or more. 1960 broken down by age and cause of Within the new cohort with a minimum death (natural or unnatural). In no age period of observation of five years, group was the number of deaths signifi mortality rose with time of exposure cantly higher than the expectancy. relative to the expectancy both for natural and unnatural deaths, and particularly for `'I initial exposures after 1960, but only be Diagnoses of fatal cancer cases came significantly higher than expectancy in the group with 5--9 years of exposure, Table 7 gives details of the 12 cases in as shown in table 5. which cancer was diagnosed as the cause 236 C c F i / a s A (' 1* 3' 4< 5 6' 7` c 31 4< 5< 6( 7f c 3 4 6 8` 3i 4t 5' 61 8' L 21 3f 4( 5< 61 T< Toibm-* rv*d years) 7 IS of death, including site of the tumor, diagnosis, date when exposure began, duration of exposure, and observation period. DISCUSSION According to preliminary reports on two animal experimental studies, a link is suggested between styrene monomer and tumors of the lymphatic and blood sys tems (7). If an association between exposure to a substance like styrene and a specific disease with a long latency like cancer is suspected, the most logical procedure would be to compare groups with sh it, intermediate and long duration of ex* posure, provided that the minimum ob servation period is at least five years from Table S. Comparison of the styrene and vinyl chloride cohorts: Relative risks of dying from specific causes, by age groups. Age (years) Styrene cohort Deaths Person-year* Expected Relative risk a b c = bf g * a/c Vinyl chloride cohort Deaths Person-years de Death rate/years f * d/e Malignant neoplasm* 30--39 40--49 50--59 80--69 70--79 X 2 5 3 1 6,693.6 4,460.6 2466.9 713.3 64.9 Cardiovascular diseases 30--39 40--49 50--59 60--69 70--79 l ,693.6 2 4,460.8 8 2466.9 14 7134 2 64.9 Other natural causes 30--39 40--49 50--59 60--69 70--79 80--69 3 4 5 5 -- -- 6,693.6 4.460.8 2466.9 713.3 64.9 -- All natural causes 30--39 40 49 50--59 60--69 70--79 80--69 5 8 21 19 3 -- 6,693.6 4,460.8 2,266.9 713.3 64.9 -- Unnatural causes 20--29 7 5450.8 30--39 3 6,693.6 40--49 4 4,460.8 30--59 3 2466.9 60--69 -- 713.3 70--79 -- 64.9 1.004 3.167 7.027 7.347 1.947 5.353 10.423 8.68 0.974 5.02 4.238 3.4 3.71 0 -- 5.957 12.49 20.855 19.687 4.888 -- 2.85 6.990 6440 2.267 1.213 0.976 0.996 0.632 0.712 0.408 0.514 --p 0.374 0.767 1.636 2.054 0.597 0.944 1.47 1448 -- -- 0.839 0.641 1.607 0.965 0.616 -- 2.456 0.429 0.63 142 -- 1 6,701 3 44204 6 14504 6 . 579.7 2 664 0.00015 0.00071 0.0031 0.0103 0.03 5 44204 0.0012 9 1,9504 0.0046 7 579.7 0.012 1 66.5 0.015 5 6,701 0.00075 4 4,2204 0.00095 3 1,950.9 0.0015 3 579.7 0.0052 2 664 0.03 2 3.12 0.64 6 6.701 0.00089 12 44204 0.0028 18 1,950.9 0.0092 16 579.7 0.965 5 664 0.075 2 3.12 0.64 3 54104 + 993.5 7 6,701 6 4420.5 2 1,950.9 1 579.7 1 664 0.00048 0.00104 0.00142 0.001 0.0017 0.015 237 ucc the beginning of exposure. The rationale The effect of a hazardous exposure of this approach is that diseases and may show more pronounced deviations deaths before a minimum period of five from expectation in the subgroups with years of observation are more likely to be long-term observation. due to previous history than to the In the case of the cohort of employees particular exposure under study. exposed to styrene, this approach has, If, on the other hand, the incidence of however, led to no significant deviation cancer increases with duration of expo from expectation. sure and if adjustment for age is properly In order to establish whether or not the considered, the evidence suggests that an results obtained for styrene exclude a association between risk substance and specific hazard, it is recommendable to disease may exist make comparisons with a group of chemi F r this reason an approach to the data cal workers that have not been exposed to analysis was made that is in accordance styrene, but are similar otherwise. with a recent criticism of the data pre The comparison shows (table 8) th't in sented by Duck et al. (2) concerning the the vinyl chloride cohort deaths occirred mortality of British workers exposed to earlier than in the styrene cohort. Cancer vinyl chloride. Several authors (1, 3, 6) mortality in the former increased with have in this context reemphasized the age, and in 12 cases out of 18 was associ't-- analysis by duration of exposure (as a ed with a period of exposure exceed mg measure of dosage), enabling the com five years; in the styrene cohort, howe /er, parison of several groups with different the period of exposure had no influ mce exposure periods. Expected deaths are on the incidence of cancer. It thus follows calculated with the total person-years that exposure to pure styrene is most experience of a subgroup, but only after probably not a cancer hazard. On the a minimum observation period (five other hand, the comparison shows that years). mortality through natural causes other Comparison of a cohort of workers than cancer was higher in the styrene employed for a long time with the popu cohort than in the vinyl chloride cohort. lation mortality for the entire time of It is thus advisable to examine the data observation would lead to the apparent in the same way in all future studies. paradox, sometimes called the "healthy The results supplemented and complet worker effect," that long-term exposure is ed the work reported on at the symposium inversely related to death. Whereas in the on "Health Aspects of Polymers" (5). reference population a number of deaths may and will have occurred in a given time, the study subjects cannot possibly have died during the timespan of this occupational exposure category for which 3? they had qualified. ACKNOWLEDGMENT litis effect, although observed for many decades, only sporadically begins to be considered appropriately in the analyses of cohort studies. As a consequence of the objection stated above, in order to adjust for this possible survival effect, the period We wish to thank Ms. Ulrike Remle, Mr. Hochadel and Mr. Kattermann for the assistance they gave us in carrying out the statistical evaluation. of five years was a first cut-off point for I the summation of person-years under observation in four categories of workers employed up to 3, 10, 13 and more than 15 years, respectively. This adjustment was a first step to REFERENCES wards a more-detailed interpretation. A minimum observation period of 10 or 15 years after the first exposure period will be considered in future studies. 1. BERRY, G. and ROSS1TER, C. E. Vinyl chloride and mortality? Lancet 2 (1976) 416. 2. DUCK, B. W,, CARTER, J. T. and COOMBES, E. J. Mortality study of work- 238 I e era in a polyvinyl chloride production plant investigations on workers employed in IS Lancet 2 (1975) 1197--1203. 3. FOX, A. J. Vinyl chloride and mortality? styrene polymerisation. Paper presented at the American Chemical Society Meeting, h Lancet 2 (1976) 416--417. Symposium on "Health Aspects of Poly 4. THIESS. A M. and FRIEDHEIM, M. Mor mers," New York, 3 April 1976. s bidity among persona employed in styrene 8. WAGONER, J. K., INFANTE, p. F. and >> production, polimerization and processing plants. Seand. j. work environ. A health 4 SAHACCI, R. Vinyl chloride and mortality? Lancet 2 (1976) 416--417. n (1978):' suppl. 2, 203--214. 7. --. MCA styrene inhalation study suggest 5. THIESS, A. M, FRIEDHEIM, M. and possible tumour risk in laboratory rats. e ROSSMANN, H. Occupational medical Cham. mark. rap. 213 (1978): 8. a o 3 1 i QUESTIONS AND ANSWERS 1 Question to Prof. THIESS ' <p Dr. HERNBERG: Prof. THIESS: Is it correct to compare a cohort of workers with a high pro portion of guest workers to the experience of the German (or local) general population. The guest workers have already been considered by the fact that the styrene cohort has also been divided into the groups "exposure before 1960" and "exposure after I960." Guest workers were em ployed only after 1960. In addition, their exposure is often short-term, and the follow-up is incomplete because they return to their native country after several years. This special group does not contrib ute much to the knowledge and will, therefore, be considered appropriately in future evaluations. (Reference: 5th Medichem congress paper: "Problems in Following-up Foreign Labour Force in Mortality Studies"). Question to Dr. RAMSEY Dr. VAINIO: Dr. RAMSEY: During the meeting we have heard that also Dow Chemicals has completed a styrene carcinogenicity study on animals. Would you like to enlighten the study further? There is nothing further that I can add to knowledge about this study that is available at this time. ....... asktvs" 239 UGC 057508 Scand. j. work environ. & health 4 (1978): suppL 2, 24(1--248 >* n Epidemiologic investigations of styrene-butadiene rubber production and reinforced plastics production by THEODORE J. MEINHARDT, RONALD J. YOUNG and RICHARD W. HARTLE 1 * MEINHARDT, T. J., YOUNG, R. J> and HARTLE, R. W. Epidemiologic investiga tions of styrene-butadiene rubber production and reinforced plastics production. Scand. j. work environ. & health 4 (1978): suppL 2, 240--248. The United States Na tional Institute lor Occupational Safety and Health (NIOSH) began an epidemiologic study of workers employed in the styrene-butadiene rubber (SBR) industry during 1978. This study was prompted by reports of relatively high numbers of leukemia deaths occurring within SBR production work populations. Simultaneous with the initiation of this Investigation, the University of North Carolina released a report associating an excess risk of death due to hematopoietic and lymphatic malignancies among workers producing several synthetic rubbers, including SBR. This report pre sents NIOSH's preliminary mortality observations and a discussion of progress made on the analyses of contaminants found in two SBR production facilities. Currently, NIOSH is determining the feasibility of doing an epidemiologic study in the rein forced plastics industry. Interest in this study developed as part of an effort to determine health hazards associated with occupational exposure to styrene. Most of the technology for the reinforced plastics industry developed in the 1990s, and there fore this process represents a relatively new industry. This report also includes in formation on environmental conditions observed in the reinforced plastics Industrv and enumerates some of the complicating characteristics of this industry which in crease the complexity of this study. The only epidemiologic studies of which I am aware that have considered the mortality patterns of workers exposed to styrene are the two studies being pre sented today by Dr. Thiess and Dr. Fischbein. These reports are concerned with the mortality patterns of individuals who have worked in industries that involve environmental exposure to low levels of 1 National Institute for Occupational Safety and Health, Division of Surveillance, Hazard Evaluations and Field Studies, IndustryWide Studies Branch, Robert A. Taft Labo ratories, Cincinnati, Ohio, U.S.A. styrene. My presentation outlines ongoing research efforts being made by the Na tional Institute for Occupational Safety and Health (NIOSH) in United States in dustries with both low and high level en vironmental exposures to styrene. Un fortunately, these efforts have not as yet progressed to the point where analyses and results are available. Potential harmful effects associated with exposure to styrene have been noted in the literature for years. It was not until leukemia cases were reported in styrene-butadiene rubber (SBR) plants in 240 .a piujwii'.'..i_ j ... m -iau -- ^^ * i. -v.1* * * v t3* ; -' A-''- . V^ .'. ` ** *\ -j Ya*\*-.* -m -.1 ----.*-?rJrws^s --* ucc 057509 the United States in March of 1976 that EPIDEMIOLOGIC INVESTIGATION IN NIOSH focused its attention on the pos THE STYRENE-BUTADIENE RUBBER sible carcinogenicity of styrene. Almost INDUSTRY simultaneously with NIOSITs decision in 1976 to study the SBR industry, a number of articles appeared concerning the pos sible mutagenicity and carcinogenicity of styrene and its metabolites. Styrene-butadiene rubber (SBR) is the most widely used synthetic rubber in the world. The United States Government foresaw shortages of natural rubber dur Styrene is known to be metabolized to styrene oxide (2), which was shown by several authors to be mutagenic in the Ames microbial test system (5, 10). Fin nish researchers reported that an uniden tified styrene metabolite was positive for ing World War II and financed construc tion of fifteen SBR plants, sixteen buta diene production facilities and five styrene plants. Between 1946 and 1955, these plants were sold to various private com panies (12). microbial mutagenesis after activation of styrene by liver microsomes (16). The mutagenesis of a styrene metabolite indi cates that styrene exposure may be related to carcinogenesis since 90 / of the carcin ogens, identified through various animal test systems, studied by the Ames tests have been shown to be mutagenic (6). In January 1976 two men employed at adjacent SBR facilities in Port Neches, Texas, died of leukemia. This informa tion, combined with knowledge of other leukemia eases at the two SBR plants, began to concern the local Oil, Chemical and Atomic Workers' Union representa tives, who in turn relayed this concern The eight malignancies of the hemato poietic and lymphatic systems among SBR workers were of particular concern in view of a previous study reporting a "statistically reliable" decrease in the amounts of blood hemoglobin, erythro cytes and thrombocytes in rubber process ing plant workers (1). Additionally, a separate preliminary epidemiologic study of workers exposed primarily to styrene to the company. Consequently, local com pany officials reviewed plant records and identified eight leukemia cases among employees within the past tew years. This apparent excess prompted the com pany officials to request that NIOSH investigate the possibility of an associa tion between employment in SBR produc tion and the development of leukemia or related malignancies. discussed what appeared to be an exces Simultaneous with NIOSH becoming sive percentage of deaths due to leukemia* aware of leukemia cases among workers and lymphomas. This latter finding must at SBR facilities, McMichael et al. (7) be tested further because the deaths were from the University of North Carolina's identified from a worker population with potential concomitant exposure to benzene and other recognized chemical hazards (4). The possibility that exposure to styrene may be associated with leukemia and other forms of cancer could represent a serious occupational health problem. In the United States alone, it is estimated that 4,082,400 t of styrene were produced in 1977 with an estimated 500,000 workers occupationally exposed to this substance. Thus, NIOSH decided to pursue further Occupational Health Studies Group re ported the cause specific mortality ex perience among a cohort of 6,678 male rubber and tire manufacturing workers and various subgroups of that total popu lation (broken down by occupational job titles). A relative risk for lymphatic and hematopoietic malignancies of 6.2 was re ported for a synthetic plant, a work area that included production of elastomers, particularly SBR. In a subsequent un published report (entitled "Toxicologic, Industrial Hygiene and Epidemiologic the question of human carcinogenicity of Considerations in the Possible Association styrene by studying employees in the re Between SBR Manufacturing and Neo inforced plastics industry where there is plasms of Lymphatic and Hematopoietic much greater potential for styrene expo Tissues") by McMichael et al. the relative sure than in the SBR industry and not risk lor hematopoietic and lymphatic sys nearly as many confounding chemicals. tems was reduced to 2.4 based on a case- 241 16 -swei --wm SwiS* Toble i. Observed leukemia initially reported among currently employed, retired or pensioned SBR workers. Case 1 2 3 4 S 6 7 8 9 Duration of employment 19 20 21 23 17 28 23 23 22 Beginning date of employment (year) 1930 1955 1938 1931 1943 1943 1943 1943 * Diagnosis based r 1 company record. Date of death (year) 1971 1978 Alive 1976 1972 1971 1978 1987 Alive Age at death Cell type* 62 Lymphocytic leukemia 46 Acute myelo genous leukemia Lymphocytic leukemia 59 Acute myelo cytic leukemia 77 Acute lympho blastic leukemia 64 Myelocytic leukemia 32 Acute myelo cytic leukemia 63 Acute myelo- blastic leukemia Chronic lympha tic leukemia Table 2. A typical SBR recipe (part per 100 parts monomer). Ingredient Quantity Butadiene Styrene n-Dodecyl mercaptan Paramenthane hydroperoxide Sodium formaldehyde sulfoxylate Soap flakes Water 73.0 23.0 0.3 0.10 0.10 3.0 180.0 control study at the same plant, but was still statistically significant. In addition to the eight leukemia cases initially identified by the companies, an other case was identified by NIOSH short ly after it initiated the study. Table 1 presents basic information on these indi viduals, as well as their specific cell type of leukemia. It is important to remember that these initially observed diagnoses are based on company records and that some of these may be assigned different Inter national Classification of Disease Codes when all of the information on the death certificates is considered. Additionally, it is known that case number five was treated for recurring respiratory infec tions with a drug which has been linked with blood dyscrasia and that case number nine had potential short-term exposure to benzene while he worked on an experi mental project. Eight of these nine cases were diagnosed after 1971. These cases were identified from company health rec ords which were available only on active, retired, or pensioned employees. There fore, the health status of terminated em ployees, the largest segment of people who have worked at the plants, remains to be evaluated. The two SBR plants in Port Neches, Texas, were selected for study so that the potential of a health risk in these two plants could be evaluated and the salient question of potential health hazards as sociated with SBR production could be addressed. The method of study chosen for this investigation was a retrospective cohort mortality design. The population, composed of approximately 5,600 workers, represents a large stable work force for which adequate personnel records are available. Later this year data collection on all members of the study population should be complete, and analysis can begin. Process and chemical hazards Styrene-butadiene rubber is a copolymer of styrene and butadiene. It is made through an emulsion polymerization proc ess. An emulsion system normally con- c T ' h Trf- ^ ' *** < ih ' n*f-i - * f. **~*-^- ..... .............- '*-*'' , 'nsV etf 11 - -* 1 -1^ - .-- . --m? _ ifcHSitsi'AsC iWO&r3$ 057511 tains water, monomers, initiator, and foam (12). An initiator system which usually contains paramenthane hydroperoxide and sodium formaldehyde sulfoxylate is used in combination with a mercaptan to poly merize butadiene and styrene. A typical SBR recipe is found in table 2. The styrene butadiene, soap, and initiator are pumped continuously from separate storage tanka into and through a series of agitated reac tors maintained at proper conditions such that the required degree of conversion is reached at the exit of the last reactor (12). The termination of the reaction is caused by addition of a "shortstop* such as so dium dimethyldithiocarbamate or sodium polysulfide. The "shortstop" reacts rapidly with free radicals and oxid'zing agents and thus destroys any remaining initiator and prevents the formation of new chains. The unreacted monomers are recovered, first the butadiene by distillation and then the styrene by steam stripping. An anti oxidant, such as a mixed alkylated diphenylamine, is added to protect the prod uct from oxidation. The latex is partially coagulated with brine and then fully co agulated with dilute sulfuric acid. The coagulated crumb is then washed, dried, and baled for shipment (12). Representative employee 8-h timeweighted average exposures to styrene and butadiene for specific job classifica tions in one of the SBR facilities are de picted in table 3. The United States threshold limit values for exposure to styrene and butadiene are 100 ppm and 1,000 ppm, respectively. It is readily seen that exposures to both chemicals are low when compared with existing United States standards. Other chemicals used in SBR produc tion have also been associated with a carcinogenic risk. Phenyl-beta-napthylamine was historically used as an antiox idant at some SBR facilities. This chemi cal has been demonstrated to be meta bolized in the body to beta-naptbylamine, a known human bladder carcinogen. Another example is the extender oils used in approximately 70 */ of SBR pro duction (11, 13). Furthermore the Oil, Chemical, and Atomic Workers' Union arranged for Kettering Laboratories to conduct skin painting experiments with five extender oils used at one of the SBR facilities in the United States. Prelimi nary results indicated a positive carcino genic response for three of these oils in the animal test system (3). Of the rec ognized potential chemical hazards used in the SBR industry, only styrene has been speculatively linked with a possible risk of hematopoietic and lymphatic malig nancies. However, it would be inappro priate at the present time to attribute the observed leukemia cases solely to the occupational exposure of styrene because of the complex chemical nature of this industry and the presence of other chemi cals which have also been associated with cancer. Table 3. SBR -- Summary plant data of employees' 8-h time-weighted average (TWA) exposure to styrene and butadiene at an SBR facility, April 1977. Styrene Job classification ,,,,_iM Range (ppm) Mean Median (high-low) (ppm) (ppm) Butadiene Range (ppm) Mean Median (high-low) (ppm) (ppm) Technical services personnel 12 Production foreman l Head production operator s Production operator 24 Operator helper 3 Pipefitter S Electrician 9 Maintenance mechanic 17 Carpenter 4 Common laborer 17 Instrument man 3 2.14--0.09 0.19-0.15 0.55--0.11 2.14--0.12 0.10--0.14 0.22--0.00 0.81--0.00 0.11--0.10 1243--0.10 1.44--0.09 0.55 0.37 0.55 114.55--0.21 19.85 0.15 0.15 0.15 1.16--1.16 1.16 0.28 0.14 0.27 69.61--0*25 15.50 0.38 0.18 0.30 33.21--043 340 0.15 0.19 0.15 1.81--048 0.79 0.13 0.10 0.13 1.78--0.18 0.74 0.11 0.10 0.11 044--0.18 042 0.17 0.12 0.10 44.30--0.19 3.15 0.11 0.11 0.11 30.60--0.20 7.80 3.00 0.61 3.12 8.22--0.17 1.52 0.85 0.41 0.84 174.14--0.17 58.62 0.63 19.68 1.16 1.18 047 15.81 0.30 3.41 0.30 0.77 0.41 0.73 040 042 0.34 340 040 7.73 0.44 1.63 1.56 58.30 243 & % Mi Sue ** dn>>y.i3j HZ&jgjr EPIDEMIOLOGIC INVESTIGATION IN THE REINFORCED PLASTICS INDUSTRY As stated earlier, it was decided by NIOSH that a second epidemiologic study focusing on workers with a high occupational ex posure to styrene and few concomitant chemical exposures should be undertaken. Three distinct industrial `settings, styrene and polystyrene manufacturing, poly styrene fabrication, and reinforced plastics production, were identified. It was de termined during plant visits in these three settings that styrene and polystyrene manufacturing and polystyrene fabrication are not labor intensive and represent low potential for worker exposure. The rein forced plastics production industry, on the other hand, has a large number of workers potentially exposed to higher con centrations of styrene. Therefore, this environment probably represents the best opportunity to evaluate qualitatively any potential health risk. The reinforced plastics industry in the United States developed during World War II with the demand for rigid materials of high strength and low weight. This in dustry has rapidly expanded from a pro duction level of 3,402 t of reinforced plastics in 1945 to an estimated 795,614 t in 1977. It has been projected that this Industry will continue to expand at an average annual growth rate of 15 */ through 1981 (14). The Reinforced Plastics and Composite Institute of the Society of plastics Industry has estimated that be tween 60,000 and 120,000 people work in the U.S. reinforced plastics industry. After the various processes used to make reinforced plastic products within the United States were evaluated, the rein forced plastics boatbuilding process was selected for study. This process was se lected because of its relatively long opera tional history compared to most reinforced plastic processes and because of the high potential for worker exposure to styrene. Method of study and efforts to identify cohort The epidemiologic method which will be used in this study is the same as that used in the study of the styrene-butadiene rubber industry -- a retrospective cohort mortality study. The reinforced plastic boatbuilding industry did not begin full operation until the mid to early 1950s and tends to be comprised of modest-sized operations (facilities with 200 to 300 em ployees or fewer). Before a statistically meaningful number of person-years at risk can be obtained, the working population from several facilities with a relatively long operational history will need to be included in the study. Additionally, these person-years at risk must have occurred long enough in the past to allow for clinical manifeststio?' of any chronic health prob lems associated with occupational ex posure. The preliminary field visits were made to the larger !>oat manufacturing facilities in the state cf Washington. Based on the information gathered at these plants, it has been determined, so far, that the workers from two facilities could be identified and included in the study popu lation. The number of production workers currently employed in one of these plants is approximately 290 and has grown con stantly since the start of the operations in 1952. The other facility began operation in the mid 1950s with a work force that has steadily expanded over the years to a current population of 485 individuals. To increase the size of the cohort, additional boatbuilding facilities are being identified for inclusion in the study. Process and chemical hazards Before the manufacturing process can be initiated, a prototype hull or deck which is made of wood, fiber glass, and resins is produced. The prototype is an exact du plicate of the reinforced fiber glass boat which is to be produced. This prototype is used to make a mold by building up layers of fiberglass doth and polyester plastic on its surface. The interior surface of the mold has the exact shape of the exterior of the boat being built. The mold is waxed on the inside before the manufacturing process starts. The manufacturing process begins with a worker spraying the inside of the 244 mold with a substance known as gelcoat, which makes up the smooth solid surface of the boat's exterior. Next, layers of fiber glass and polyester resins are applied by spray gun and by hand. The reinforced fiber glass is rolled, brushed, and squee geed to work out airpockets and other imperfections. At this point the structure of the hull and deck has been formed. The resin and fiber glass are cured while on the mold for a prescribed amount of time, and then the deck or hull is taken off the mold and is ready for finishing (8). The following description outlines the chemistry of the reinforced plastic process. The basic reaction is the polymerization of an organic acid and alcohol to form an ester. Styrene monomer is used in the reaction as a solvent and a reactant (8). It is a solvent because it dilutes the add and alcohol and thereby keeps them in solution until mixed with a catalyst Styrene is also a reactant because the soluble polymers formed by the add and the alcohols are cross-linked (Le., con nected by siyrene) to produce polymers which are insoluble in most common solvents. Prior to the formation of these insoluble polymers, the chemicals are mixed with glass fibers so that the desired end product is produced (15). The major chemical hazard resulting from the handling and application of rein forced plastics is assodated with sytrene. Styrene comprises 40 / or more of the gelcoat and the resin, and is a volatile substance under normal conditions (9). Exposure to styrene can occur through several different routes: inhalation of atomized mist from the spray gun, dermal absorption during the hand work-up of the resin and fiber glass, and inhalation of styrene vapor evaporated during the curing process. In my conduding remarks I would like to mention several additional ongoing epidemiologic investigations which may be pertinent: 1. In the United States the Society for the Plastics Industry (SPI) is conducting a cross-sectional medical study on approxi mately 300 active and a few retired em ployees at a reinforced plastic boat plant2 2. SPI is also considering the feasibility of having a mortality study conducted in various segments of the reinforced plastics industry. 3. In addition Dr. Ahlmark from Swe den is currently conducting mortality and carcinogenic assessments of reinforced plastics workers from a number of plants. The assessments are based on national record systems and cancer registries in Sweden. At this point, NIOSH would like to solicit any eommen s or suggestions on how these ongoing studies or new epide miologic studies could better address the important questions nised so far at this conference. NIOSH i' also interested in learning about any firther epidemiologic research being conduc ted by governmental agencies, labor associations, or private in dustries in other countries. REFERENCES 1. ALTMONA, S. T. Occupational hygiene and conditions of workers' health in prepara tion shops of fine factories. Gig. tr. prof. zabot. 15 (1971): 4, 24--28. 2. LEIBMAN, K. C. and ORTIZ, E. Epoxide intermediates in microsomal oxidation of olefins to glycols. J. pharmocol. exp. ther. 173 (1970) 242--24*. 3. LIFELINES, OCAW HEALTH AND SAFE TY NEWS. Update on leukemia investiga tion among SBR workers. Lifelines, OCAW health and safety newt 24 (1377): 4. 4. LIUS, R. and NICHOLSON, W. J. Cancer experience among workers producing sty rene monomer. In: Proceedings of NIOSH styrene-'butadiene briefing, April 30, 1970. 5. LOPRIENO, N., ABBONDANDOLO, A, BARALE. R., BARONCELLI, S., BONNATI, S.. BRONZETTI, G.. CAMMELLINT, A, CORSI, C., CORTI, G., FREZZA D., LEPORINE C., MAZZACCARO, A, NIERI, R., ROSELI.INI, D. and ROSSI, A. M. Mutagenicity of Industrial compounds: Vinyl chloride, styrene and their possible metabolites. Mutat. ret. 38 (1978) 114--113. 8. MC CANN, J. Detection of carcinogens as mutagens in Salmonella microsomal test Proc. not. oced. sci. U4A 73 (1978) 950--954. 7. MC MICHAEL, A J,, SPERTAS, R., GAM BLE, J. F. and TOUSEY. P. M. Mortality among rubber workers: Relationship to specific jobs. J. occup. med. 18 (1976): 3, 178--183. 245 \ Ss r= : \65- ! & & ucc 057514 V. 8. MILLER, S. Manufacturing process de scriptions (reinforced plastic boats). Glastron Boat Company, Austin, Texas 1972, pp. 1--3. 9. MILLER, S. Manufacturing process descrip tions (reinforced plastic boats). Glastron Boat Company, Austin, Texas 1972, p. 4. 10. MILVY, P. and GARRO, A. J. Mutagenic activity of styrene oxide (1,2 epoxyethyl benzene), a presumed styrene metabolite. Mutat. res. 40 (1976) 15--18. 11. SALTMAN, W. M. Butadiene polymers. In: Encvclopcdia of polymer science and tech- notopy, Interscience Publishers 1965. 12. SALTMAN, W. M. Styrene-butadiene rub bers. In; M. MERTON (ed.), Rubber technology. Van Noetrand Reinhold Com pany, New York, N. Y. 1973, pp. 178--198. 13. SHREVE, R. M. Chemical process indus tries (3rd ed.). McGraw-Hill Book Compa ny, New York, N.Y. 1967. 14. SOCIETY OP THE PLASTICS INDUSTRY INC., REINFORCED PLASTICS/COM POSITES INSTITUTE Guidelines for man. ufacture of reinforced plastics/composites. New York, N.Y. 1977, pp. 1--3. 15. SOCIETY OP THE PLASTICS INDUSTRY, INC, REINFORCED PLASTICS/COM POSITES INSTITUTE. Guidelines for manufacture of reinforced plastics/compoeitcs. New York, N.Y. 1977, pp. 6--7. 16. VAINIO, H, PAAXKONEN, R, RONNHOLN, K, RAUNIO V, and FELKONEN, O. A study of the mutagenic activity of styrene and styrene oxide. Scand. f. work environ. A health 2 (1976) 147--151. QUESTIONS AND ANSWERS Question to Mr. ME3NHARDT and Dr. FISCHBEIN Dr. A. LEHTNIEMI: Mr. ME1NHARDT: Dr. FISCHBEIN: 246 Upper respiratory tract infections are more common in employees exposed to various chemicals in the rubber and plastic industries, including *tyrene-butadiene-acrylonitrile; on the other band, viral etiology has been proposed for human leukemia. Have you taken into account the frequency and typo of respiratory infections among those workers who died to leukemia or lym phoma? In other words, the high incidence in leukemia could be an indirect effect related to irritative properties and not to mut agenic properties of styrene, (all "flu" will never be recorded!) In the ongoing NIOSH epidemiologic study of SBR workers, this hypothesis has not been considered to date, but it is very interest ing. Respiratory health hazards in this industry are of interest, and the analysis of mortality patterns in this investigation will pay specific attention to respiratory problems. The causal relationship between respiratory infections and leuke mia is well known, since patients with leukemia have a decreased resistance to infectious diseases in general. This may be due to Impaired function of leucocytes or impairment of immunological competence. This is a post facto phenomenon, however. In our clinical, cross-sectional study there was a higher prev alence of respiratory symptoms among those with high exposure than among the workers with low exposure. In addition there was a trend towards a higher prevalence of pulmonary function abnormalities (FEVj/FVC < 0.73) among workers with higher urinary mandelic add excretion. With regard to the mortality study I have no information about the Incidence of respiratory infections among the indi viduals who subsequently developed leukemia. It seems that It would be very difficult to obtain accurate data on this matter In a retrospective-prospective study. Generally, there was a very low mortality in respiratory disease in the mortality study pre sented (1 case observed against 7 expected). . .-on.,44 4 `--`W*'-.-,? -- i q i*ibf * - , --Si?TM* --... t. _____ ______-- ,, --rrrTvia-"f1'~ ,`v" ----------------- -- ** -f -r .sti-a--i r:*li~o4trsinWi -- > ' ' --------------- -- MM* U&7515 Seand. j. work environ. A health 4 (1978): suppL 2, 247--252 Mortality experience of styrene-polystyrene polymerization workers1 Initial findings by WILLIAM J. NICHOLSON,* IRVING J. SEUKOFF * and HERBERT SEIDMAN NICHOLSON, W. J., SEUKOFF, L J. and SEIDMAN, H. Mortality experience of styrene-polystyrene polymerization workers: Initial findings. Seand. j. work environ, A health 4 (1978): suppL 2, 247--252. A group of 580 individuals has been identified as employed on 1 May 1960 with at least five years seniority in a plant manufactur ing styrene and polystyrene. In this plant workplace exposures included styrene, benzene and ethylbenzene, among other materials. All of the 560 individuals have been traced through 1975 and their vital status determined. Expected and observed deaths, by cause, were determined from 1 Ma; 1960 or the tenth anniversary of employment in the plant through 31 December 1975. Over this relatively short period of time a deficit of deaths compared to that of the general population was observed (106.41 expected versus 83 observed). Among the 83 deaths, one was of leukemia, one of lymphoma and an additional death was accompanied by leukemia. A review of 361 additional death certificates revealed five additional cases of leukemia and four of lymphoma. The available informations from the limited follow-up in time of the cohort and from the randomly collected death certificates, while suggestive of a possible risk, is not definitive. Key word*: benzene, leukemia, lymphoma, mortality, polystyrene, styrene. After the recognition in 1974 that exposure to vinyl chloride posed a significant human carcinogenic risk, producing liver hemangiosarcoma and the other malignant tu mors (3, 5, 6), attention was directed to other widely used plastics in which mono mer exposure to workers or the general population could be of significance. Among such monomers was styrene (vinyl ben zene), widely used in the production of polystyrene, which has a chemical struc ture with similarities to that of vinyl chloride. Of additional concern was the i This paper was presented by Dr. Alf Flschbein. * Environmental Sciences Laboratory, Depart ment of Community Medicine, Mount Sinai School of Medicine, New York, N.Y., U.S.A. s American Cancer Society, New York. N.Y-, U.S.A. possibility that worker exposure to ben zene, a leukemogenic agent, could occur in the production and purification of styrene monomer. Furthermore, a potential for styrene exposure to the general population exists, albeit at considerably lower concen trations than those of the workplace. Such exposures could result from the leaching of unreacted styrene from food, and bever age containers and from the combustion of polystyrene products. This latter source is of importance because polystyrene decom poses under heat to its monomer. More recently, data have been presented that suggest styrene may be carcinogenic, producing leukemias and lymphomas (1, 4). This possibility follows the finding in 1976 of eight eases of leukemia in two styrene-butadiene production facilities in Texas, and a stated increase in deaths from 247 leukemia and lymphoma in an Ohio styrene-butadiene rubber manufacturing plant Of the eight cases six wer among a group of 146 known deaths and two were under treatment or in remission. For the investigation of whether health effects can be associated with styrene pro duction and polymerization, a clinical survey and a mortality study of a large monomer and polymerization plant was conducted in 1975. The findings of the clinical survey appear elsewhere in these proceedings. Here we report the results of a mortality investigation of 560 long term workmen employed at the facility. MATERIALS AND METHODS Among other information, a seniority list, as of 1 May 1960, was made available by Table I. Distribution of ages and times of em ployment of 380 individuals at a styrene pro duction and polymerization facility as of l May 1960. (years) Years since onset of work 5*9 10--14 15--19 All years 20--29 30--39 40--19 50--39 60--69 >70 All ages St 4 81 109 72 84 33 17 81 2 55 12 202 41 177 40 92 22 31 i3 247 197 116 560 Table 2. Major work activity of 560 individuals during employment at a styrene production and polymerization facility. Type of work Number High exposure categories Styrene production Polystyrene polymerization and extrusion Development and special products Maintenance Low txpoiuTt categoric* Service and utilities 56 129 53 206 116 Local 8--74 of the Oil Chemical and Atom ic Workers' Union, which represented em ployees at the facility, The list was com prised of 711 individuals, 563 of whom had five years of employment in the plant. These latter were considered for inclusion in a cohort in which each individual would be followed prospectively from 1 May 1960 or upon attaining his 10th anniversary of employment The five-year work criteria was applied so that the group would have significant exposure to styrene, and the 10-year criteria was applied so that long term health effects, such as cancer, could also be considered. Of the 563 individuals, three were eliminated, one because he served as a union representative beginning in 1948 and was not working in the plant on 1 May 1960, one because of death prior to achieving his tenth anniversary of em ployment, and one, who was the sole fe male in the group, because different mor tality statistics would be required. Table 1 lists the breakdown according to seniority and age of the 560 individuals. Table 2 categorizes them according to ma jor employment activities within the plant The categorization of the departments ac cording to relatively high or low exposure was made on the basis of air concentrations measured in a Health Hazard Evaluation during 1974 by the National Institute for Occupational Safety and Health (NIOSH), by worker descriptions, and by body bur den of styrene metabolites measured during the course of the previously men tioned clinical survey. The air concentra tions of styrene generally ranged from 5 to 20 ppm in the "high" exposure areas and less than 1 ppm in the "low" areas at the time of the NIOSH survey. However, wide excursions from these ranges oc curred at specific locations. A good corre lation of the presence of styrene metabo lites with exposure classification is shown in the paper by Lorimer et al. (2) in these proceedings. Furthermore, while group exposures generally applied, there could be some individuals within each group who had a long-term exposure different from that of the group as a whole. The current production operations are shown in fig. 1. Ethylbenzene is received at the plant in tank cars and reacted with superheated steam and a catalyst of iron 248 ***** ^yiniir ri--yin*inif*ii>flfUlfi t **^*-*,?*--'- - * . ii*. ' v; Shiftyjiffifr nOoUcTtSitIOctM THAT* MOOUCTMSIOSCOm Fig. 1. Current operations and distribution o workmen in a styrenepolystyrene production facility. oxi- e to produce styrene monomer. The crude styrene so produced is purified with the removal of unreacted ethylbenzene, benzene, toluene and xylene. These impu rities are hydrogenated in another opera tion and distilled to separate the compo nents for later use or distribution. Prior to 1963, benzene was produced at the plant from the by-products of coke ovens and reacted to form ethylbenzene. In the polystyrene production, the puri fied styrene is reacted with water, cata lysts, and various agents to produce poly styrene beads. These may later be impreg nated with pentane in a closed system to form an expanded polymer. Alternatively, they may be extruded and formed into sheets. Other activities in the plant include the production of butadiene-styrene latex and pilot plant operations for research and development work on styrene polymers specific to a customer's specifications. A large maintenance force is employed at the plant and includes such trades as elec tricians, laborers, welders, insulators, and others. These men work throughout the plant as required, often at times of me chanical breakdown or leakage of various systems. Thus, exposure of individuals in this category could often be high. The utilities and service employees include those working in the power house and janitorial or other service jobs. Their ex posures would be generally of a lower category. RESULTS All of the 560 individuals had been traced as of 31 December 1979, and their vital status determined. At that time, 233 were still employed at the facility, 83 were de ceased, and 241 bad retired or terminated employment. Each of this last group w re contacted directly for verification as a member of the cohort and for information on work adtivities. For the 83 deceased, certificates of death were obtained in all cases. Additionally, when appropriate, clinical information, radiographs, and pathological specimens were sought from hospitals. Autopsy protocols were received for 18 cases and other information for 13. Table 3. Expected and observed mortality ex periences of 560 individuals employed at a styrene production and polymerization facility prior to 1 May 1955, followed 10 years after onset of exposure (1 May 1960 -- 31 December 1975). Cause of Death Expected Observed All causes of death Cancer Cancer of the lung Leukemia Lymphomas Other cancer Heart and circulatory dimiM Respiratory diseases Other causes of death 106.41 21.01 6.99 0.79 1.25 11.98 56.35 5.64 22.41 83 17 6 1 1 9 52 1 13 249 17 A review of these additional data revealed that one individual, listed as deceased from pancreatic cancer on the certificate of death, actually died from acute pancrea titis. Table 3 shows the expected and ob served mortality experience, as verified for the 560 individuals from 1 May 1960 through 31 December 19757 The expected rates of death were calculated with a mod ified life table analysis and data on the mortality experience of the general popul ation of the United States. The results in dicate that about 79 */ as many deaths as expected occurred during 'this perod of time. Deficits of this order are commonly found in studies comparing the mortality experience of working groups with that of the general population. This deficit results, in part, because groups of employed work men are healthier than a corresponding age group in the general population which would include individuals terminally ill and others unable to hold jobs because of disability. This "healthy worker effect" can influence total mortality data for ten or more years and that of cancer for two to five years. While the data are limited because of small numbers, some infor mation on these expected time effects can be seen in tables 4 and 5, where the cohort mortality experience in different calendar years indicates that death rates more closely approach those of the general pop ulation as time from identification of the cohort increases. This change is also seen in an analysis of mortality patterns ac cording to time from first exposure where the highest ratios of observed to expected deaths are found in the groups having more than 20 years since first employment An analysis of mortality experience by extent of exposure is shown in table 6, where maintenance and production workers are compared to utility and service employees. No particular patterns were identified in this separate mortality analysis, but again the data were limited by small numbers. Because of the high exposure to ben zene experienced by some individuals in the cohort during the 1943--1962 period of time, and to a lesser extent thereafter, special attention was paid to leukemia as "'~+*A ' .` -.T+tja+i ffnwtt '.5#T.K Table 4. Expected and observed mortality experiences of 960 styrene production and polymeri zation workmen in three calendar period* of time. Cause of death 1960--196* Ex pected Observed 196*--1970 Ex pected Observed 1971--1975 Ex pected Observed All causes of death Cancer Heart and circulatory diseases Respiratory diseases Other 26.76 4.78 14.01 1.94 6.43 14 2 8 0 4 39.68 6.87 18.42 2.23 8.14 28 6 18 0 4 43.97 9.36 23.92 2.89 7.84 41 9 26 1 5 Table 5. Expected and observed mortality experiences of 960 styrene production and polymeri zation workmen according to yean from onset of employment. Cause of death 10--19 Ex pected Observed 20--29 Ex pected Observed >30 pSid -Ob*"TM1 All causes of death Cancer Heart and circulatory dlseaaes Respiratory disease* Other 39.79 7.29 20.06 2.30 10.17 20 2 11 0 7 99.94 1243 3244 3.94 11.33 99 15 37 1 6 6.79 4 1.43 0 349 4 0.90 0 0.91 0 250 - ------- ' o--- 057519 fttf? - a. ::ed m iort dar ore op- the een acere `.ed ing nt. bv 6. .on .nd .ns :tv ed -n- :n of ir as Table 6. Expected and observed mortality experiences of 560 styrene production and polymeri zation workmen according to work activity. Cause of death All causes of death Cancer Heart and circulatory diseases Respiratory diseases Other Production and polymerization Ex pected Observed Maintenance Ex pected Observed Utilities service Ex pected Observed 40.26 8.17 20.36 2.41 9.12 31 4 20 0 7 42.49 8.03 23.22 2.74 8.48 36 10 23 0 3 23.66 4.79 12.57 1.49 4.81 16 3 9 1 3 a cause of death. While one death ,n the cohort was directly attributed to leukemia, a second individual, deceased of ' coro nary, also had leukemia at the time of death. During the course of this study, an additional 21 death certificates of individ uals who died in recent years were made available to us from union files. These were of individuals that did not have five years of employment as of 1960 and were thus not included in the cohort under con sideration. Among these 21 deaths, there was an additional leukemia and an addi tional lymphoma. Thus, in 104 deaths, there were two of leukemia, one with leu kemia and two lymphomas. While the total number is small, the finding of five leu kemias and lymphomas among 104 deaths suggested that further investigations of the plant mortality experience would be use ful. This information was provided to the company and the union representing the plant workers. Following its receipt, the company initiated an epidemiologic study of all individuals employed for six months or more at that particular plant and sup plied information on its progress to NIOSH. From information on 444 death certificates it was determined that seven deaths in volved leukemia and an additional five, diseases of the lymph system. The work histories and exposures of the 444 indi viduals, however, were not available. The breakdown was as follows: Acute lymphoblastic leukemia Acute blastic leukemia Acute myelogenous leukemia Chronic lymphatic leukemia Chronic myelogenous leukemia 1 1 3 l 1 Hodgkins disease Lymphosarcoma Retroperitoneal lymphoma 3 1 1 Unfortunately, the small numbers and limited follow-up in time of the cohort does not provide definitive information on the risk of death from leukemia or lympho ma in this plant (other than to indicate that it was not extraordinary). Similarly, available information from the randomly collected death certificates, while sugges tive of an excess risk, is not definitive. More extensive epidemiologic studies, over longer time periods, will be required. Such studies, however, must take into account the possible exposures to benzene in the consideration of the etiology of leukemia. REFERENCES 1. LEMEN, R. A. and YOUNG, R. Investi gations of health hazards in SBR facili ties. Proceedings of the NIOSH styrenebutadiene briefing, Covington, Kentucky, April 30, 1976 (HEW publ. no. (NIOSH) 77--129). 1976, pp. 3--6. 2. LORIMER, W. V., ULIS, R., FISCHBEIN, A, DAUM, ANDERSON, H., WOLFF, M. S. and SEXJKOFF, L J. Health status of styrene-polystyrene polymerization workers. ScantL j. work environ. St health 4 (1978): suppl. 2, 220--226. 3. NICHOLSON, W. J., HAMMOND, E. C., SEIDMAN. H. and SELIKOFF, I. J. Mortal ity experience of a cohort of vinyl chloridepolyvinyl chloride workers. Ann. n.y. acad. sci. 246 (1973) 223--230. 251 WSrHwefiNr ucc 057520 4. SPIRTAS, H. Mortality among rubber workers: Jobs with SBR exposure. In: Proceedings of the NIOSH styrene-buta diene briefing, Covington, Kentucky, April 30, 1976 (HEW pubL no. (NIOSH) 77--1291 197*. pp. 9--18. 5. TABERSHAW, I. R. and GAFETY, W. R. Mortality study of workers In the manu facture of vinyl chloride and its polymers. J. occup. med. 18 (1974) 509--518. 6. WAXWEILER, R. J., STRINGER, W., WAGONER, J. K. and JONES, J. Neoplastic risk among workers exposed to vinyl chlo ride. Ann. n.y. acad, sci. 271 (1976) 40--48. QUESTIONS AND ANSWERS Question to: Dr. ALT FISCHBEIN Dr. RANTANEN: Dr. FISCHBEIN: Prof. THIESS: Dr. HILDEBRAND: We can expect that styrene is a weak carcinogen. Your exposure time was about 20--25 years and latency time maximally 20--25 years. Could we expect that we now are registering only latency period and 10 to 20 yean more are needed to show excess malig nancy frequencies? Ihe period of clinical latency is an important consideration in the evaluation of environmentally induced malignant disease. For certain asbestos related tumors, pleural and peritoneal meso thelioma, for example, the latency period may be very long and the turnon usually do not become clinically manifest until 30--40 yean after onset of exposure. The mortality study presented did. not give any conclusive evidence because of the relatively young population and short latency period. Conclusive information is urgently needed, however, in order to protect those currently exposed. Methodological approaches other than mortality studies may also contribute to the speedy solution of this problem. Interaction and synergism between styrene and other chemi cals in the work environment should also be taken into account Both carcinogenic potential and latency period may be affected by such factors. Comments on the statement of Dr, Hantanen in which he to placed, styrene as a weak carcinogen. From my point of view there isn't any evidence that allows us to call styrene a weak carcinogen at this moment. We have experiences since 1931; that means 46 years. We should avoid that unskilled people will take any action due to this remark of Dr. Rantanen. Our duty is to inform people but not to raise fears. I would like to put the remark of Dr. Rantanen into the correct perspective. You only could assume styrene to be a weak car cinogen, but up to now there is no experimental evidence for that. This is based on three other major studies which art now complete (NCJ, MCA, MALTONI) and the results, though not all conclusive, do not show any clear evidence for styrene to be a clearcut carcinogen, not even a weak one. 252 , .- . ' ".* . -X___ . *-- r Jew,.i W.`*.fMwvmkr* 411V' -*ftJ UCC 057521